Respiration-averaged CT versus standard CT attenuation map for correction of ¹⁸F-Sodium Fluoride uptake in coronary atherosclerotic lesions on hybrid PET/CT

Abstract

Background

To evaluate the impact of respiratory-averaged computed tomography attenuation correction (RACTAC) compared to standard single-phase computed tomography attenuation correction (CTAC) map, on the quantitative measures of coronary atherosclerotic lesions of ¹⁸F-sodium fluoride (¹⁸F-NaF) uptake in hybrid positron emission tomography and computed tomography (PET/CT).

Methods

This study comprised 23 patients who underwent ¹⁸F-NaF coronary PET in a hybrid PET/CT system. All patients had a standard single-phase CTAC obtained during free-breathing and a 4D cine-CT scan. From the cine-CT acquisition, RACTAC maps were obtained by averaging all images acquired over 5 seconds. PET reconstructions using either CTAC or RACTAC were compared. The quantitative impact of employing RACTAC was assessed using maximum target-to-background (TBR_MAX) and coronary microcalcification activity (CMA). Statistical differences were analyzed using reproducibility coefficients and Bland-Altman plots.

Results

In 23 patients, we evaluated 34 coronary lesions using CTAC and RACTAC reconstructions. There was good agreement between CTAC and RACTAC for TBR_MAX (median [Interquartile range]): CTAC= 1.65[1.23–2.38], RACTAC= 1.63[1.23–2.33], p=0.55), with coefficient of reproducibility of 0.18, and CMA: CTAC= 0.10 [0–1.0], RACTAC= 0.15[0–1.03], p=0.55 with coefficient of reproducibility of 0.17

Conclusion

Respiratory-averaged and standard single-phase attenuation correction maps provide similar and reproducible methods of quantifying coronary ¹⁸F-NaF uptake on PET/CT.

Introduction

Hybrid positron emission tomography and computed tomography (PET/CT) imaging with ¹⁸F-sodium fluoride (¹⁸F-NaF) has been successfully employed for the assessment of atherosclerotic disease activity in the coronary arteries and can potentially identify high-risk plaques (1–4). Currently, hybrid ¹⁸F-NaF PET/CT scans are obtained utilizing a 30-min PET acquisition in listmode format, followed by PET image reconstructions employing four or ten cardiac phases with either breath-hold or free-breathing CT attenuation correction (CTAC) maps (5–7). Previous studies on myocardial perfusion and viability have shown that misalignment of the PET emission data and the CTAC maps pose risks of false-positive findings in the clinical setting (8, 9), with loss in the predictive value if not corrected for motion (10). To ameliorate this problem, realignment of the CTAC maps and PET emission data before PET image reconstruction is sometimes required to obtain optimal predictive value (11). However, single-phase CTAC maps affect the quantitative accuracy of myocardial viability studies owing to the continuous respiratory translations during the PET emission acquisitions(12–15). Previous studies employing respiratory averaged CTAC maps (RACTAC), obtained from cine-CT scans can improve the quantitative accuracy in myocardial perfusion and viability studies(16, 17). Recently, ¹⁸F-NaF PET has been established as a non-invasive imaging modality to identify high-risk and ruptured coronary atherosclerotic plaques(1, 3, 18, 19). Quantitative accuracy and reproducibility of imaging this tracer will be critical to establish its clinical value. However, the quantitative accuracy of lesion micro-calcification activity assessed with ¹⁸F-NaF PET and RACTAC for attenuation correction has not yet been evaluated.

In this study, we aimed to evaluate the quantitative impact of using RACTAC on ¹⁸F-NaF uptake measurements, and to compare RACTAC and CTAC maps for the reconstruction of PET images (Figure 1). To this end, we aimed to compare the quantification of individual lesions using target-to-background ratios (TBR) and whole-vessel and entire coronary tree microcalcification burden using coronary microcalcification activity (CMA) assessments (2, 20, 21).

Figure 1.

Standard attenuation map-CTAC (left) vs respiratory average CT attenuation maps-RACTAC (right). Coronal (top) and transverse (bottom) images.

Materials and Methods

Study population

Twenty-three patients underwent hybrid ¹⁸F-NaF PET/CT examinations of the coronary arteries as part of the ongoing Effect of Evolocumab on Coronary Artery Plaque Volume and Composition by CCTA and Microcalcification by ¹⁸F-NaF PET study (NCT03689946)(22). Patient demographics are shown in Table 1. Inclusion in the study required a Coronary Computed Tomography Angiography (CCTA) defined high non-calcified coronary artery plaque volume (>440 mm³). Exclusion criteria were as follows: renal dysfunction (serum creatinine > 1.5 mg/dL) prior to imaging, history of allergy to iodine contrast agents, allergy to evolocumab, women who are breastfeeding, active atrial fibrillation, history of coronary artery bypass graft surgery , inability to lie flat, inability or unwilling to give informed consent, major illness or life expectancy <1 year, planned coronary revascularization or major non-cardiac surgery in the next 12 months, previous or current evolocumab use.

Table 1.

Age (SD)	66±10
Sex (Males)	21 (78%)
BMI (SD)	27±4
Hyperlipidemia	24 (89%)
Hypertension	14 (52%)
Diabetes	5 (18%)
Smoker/ex-smoker	7 (26%)
Total Plaque Volume (mm3)	837 [620–1066]
Total NCP Volume (mm3)	711 [55—859]
Total Calcified Volume (mm3)	103 [43–205]

Continuous variables reported as mean ± SD or median [interquartile range]; categorical variables reported as n (%), BMI: Body mass index, NCP: Non-calcified plaque

Imaging protocol

CCTA acquisition

The CCTA was acquired the same day as the PET scan for 13 patients, while the remaining cases (n=10) had the CCTA within 21 days (range=2–21 days). All patients were scanned with arms positioned above the head in a 192-slice Somatom Force Dual Source mCT system (Siemens Healthineers, Knoxville, TN, USA). The CCTA imaging parameters included prospective gating, 250 ms rotation time, body-mass index (BMI) dependent voltage (<25 kg/m², 100 kV; ≥25 kg/m², 120 kV), and tube-current time product of 160–245 mAs. Patients were administered beta-blockers (orally or intravenously) to achieve a target heart rate of <60 beats/min, followed by a BMI-dependent bolus-injection of contrast media (400 mg/mL) with a flow of 5–6 mL/s after determining the appropriate trigger delay defined by a test bolus of 20 mL of contrast material.

PET acquisition

Following the CCTA acquisition (at the same day or different day within a 21 days period as described above), all patients underwent ¹⁸F-NaF positron emission tomography (PET), with arms positioned above the head on a hybrid PET-CT scanner (Discovery 710, GE Healthcare, Milwaukee, WI, USA). Prior to imaging, subjects were administered with a target dose of 250 MBq of ¹⁸F-NaF and rested in a quiet environment for 180 min (Figure 2). All patients underwent 30 min PET emission acquisitions, using a standard 3-lead electrocardiogram using a free-breathing protocol, where the patients were advised to relax.

Figure 2.

Imaging protocol for PET acquistion. All patients were scanned with arms positioned above the head.

AP: anteroposterior, CTAC: coronary tomography attenuation correction, PET: positron emission tomography

Cine-CT and standard CT attenuation correction acquisitions.

Prior to the PET scans, an anteroposterior scout at 100 kV and 10 mA was obtained starting from the pulmonary artery bifurcation and extending to the diaphragm. Using the anatomical information, 4D cine-CT data were acquired with the following settings: 120 kV, 10 mA, slice thickness 2.5 mm, 0.8 sec/rotation, axial coverage of 14 cm, 7 cine acquisitions, and cine duration of 5 sec. Simultaneously with the cine-CT scan, we acquired information on the respiratory signal and its periodicity to perform quality-checks on the respiratory averaged CTAC maps using the Varian RPM system (Varian Medical Systems). In addition to the cine-CT scan, a standard AC map was acquired at the following settings: 100 kV, 40 mA, 0.5 second/rotation, 1.375 pitch factor, 5-mm slice thickness with an axial field of view of 15.4 cm, and acquisition time of 1.5 sec. The total radiation doses of 4D cine-CT and standard AC maps were 1.4 mSv and 0.8 mSv, respectively.

RACTAC map.

From the CINE CT acquisition, we created RACTAC maps from averaging the CT images acquired over 5 seconds (Figure 1). To compensate for the reduced field-of-view, the RACTAC (axial coverage = 14 cm) was patched with the standard CT AC map (axial coverage = 15.4cm) for the missing 1.4 cm. In addition, the RACTAC maps were interpolated to obtain the same resolution in the axial direction as the standard CTAC map to preserve accurate attenuation correction of the PET images. Both the CTAC and the RACTAC maps were used for attenuation correction using a vendor-provided reconstruction toolbox.

PET reconstruction.

The acquired electrocardiography-gated raw data (list mode dataset) were reconstructed using a standard ordered expectation-maximization algorithm with time of flight, and resolution recovery using 4 cardiac bins. All images were reconstructed using a 256 × 256 matrix (47 slices) employing 4 iterations and 24 subsets, followed by a 5-mm Gaussian post-filtering.

Two series of PET images, one employing CTAC and another with RACTAC, were reconstructed into using a vendor-provided software (REGRECON-REL5, General Electric, Wisconsin, USA).

Post reconstruction Cardiac motion correction

Cardiac motion-corrected images were obtained from the gated PET reconstructions through PET-PET image co-registration using a diffeomorphic registration(23). This technique enables alignment of all gates to the end-diastolic position and as a result, allows for inclusion of all PET counts acquired. We used dedicated coronary PET imaging software for image analysis (FusionQuant, Cedars-Sinai Medical Center)(18).

PET quantification

Background blood pool clearance correction

To minimize the impact of variations in background blood pool activity introduced by variations in the injection-to-scan delays (6), we standardized the background blood pool activity to an injection-to-scan delay of 180 minutes using a previously described correction factor (6) (Eq. 1)

$$\text{SUVBackground}\text{corrected}=\text{ SUVBackground}*\text{e}\left(-0.004*\left(180-\text{t}\right)\right)$$ (1)

where t represents the injection-to-scan delay in minutes.

TBR_MAX quantification

TBR_MAX was obtained using a previously described protocol (2). In brief, for individual atherosclerotic lesions, the ¹⁸F-NaF uptake was evaluated employing maximum standardized uptake values (SUV_max) obtained within 3D spherical volumes of interest (VOI) (radius 5 mm) placed on the lesions with ¹⁸F-NaF focal uptake. TBR_MAX values were calculated by normalizing the atherosclerotic SUV_MAX values by the injection-to-scan delay corrected blood pool activity measured in the right atrium (cylindrical volume of interest radius 10 mm and thickness 5 mm) at the level of the right coronary artery ostium. Lesion SUVmax and TBRmax were evaluated using the same VOIs for both reconstructions, using VOIs inserted on PET images reconstructed using CTAC as reference.

CMA quantification

To obtain the CMA(2) values, two distinct steps were performed. First, we selected the proximal and distal end of the vessel (>2 mm) and applied a vessel tracking algorithm to extract whole-vessel tubular 3D volumes of interest from CCTA using dedicated semi-automated Autoplaque software (Cedars-Sinai Medical Center, Los Angeles, CA)(24). In a tubular VOI, along the extracted centerlines, with 4-mm diameter, we measured the coronary microcalcification activity (CMA) on the PET/CT co-registered images. CMA was defined as the average SUV within the activity volume above a threshold established as the mean background SUV +2 standard deviations. The background activity was measured in the right atrium. In order to evaluate the total uptake, we added the CMA activity of all epicardial vessels (CMAtotal).

Offset calculation between CTAC and RACTAC

Offsets obtained for the two attenuation correction maps were measured in 3D, using the aortic valve as a point of reference for the translations observed in the heart. All offsets were calculated using the standard CTAC maps as the reference, serving as a gold standard for the assessments.

Statistical analysis

The data were tested for normality using the Shapiro–Wilk test. Statistical analysis was performed using MedCalc Statistical Software version 19.1.7 (MedCalc Software bv, Ostend, Belgium). Continuous, normally distributed variables were presented as mean ±SD (standard deviation), whereas non-normally distributed continuous data were presented as median [range]. We assessed CMA and TBR_MAX, using descriptive statistics and Bland–Altman plots, as well as coefficients of reproducibility. Boxplots were designed using R 3.5.0 and statistical significance of the difference between the correlated variances was calculated using the Pitman Morgan test; a two-sided p value <0.05 was considered significant.

Results

Twenty-three patients underwent hybrid ¹⁸F-NaF PET/CT examinations of the coronary arteries. A total of 34 coronary lesions were identified on reconstructions on both PET corrected with CTAC and RACTAC maps (Figure 3). Across all patients, the median (IQR) offset for the CTAC and RACTAC maps were calculated to be 5.4 mm (2.2 mm – 9.9mm).

Figure 3.

Left anterior descending artery ¹⁸F-NaF uptake. Left panel shows fused CTA and PET images reconstructed using the CTAC maps, whereas right shows the same CTA fused with PET an image reconstructed using RACTAC. There is good agreement between the measurements.

CTA = computed tomography angiography, CTAC = Computed Tomography Attenuation Correction and RACTAC = Respiratory averaged computed tomography attenuation correction.

Per-lesion analyses (TBR)

TBR values obtained using the same delineation of the lesions for reconstructions employing CTAC and RACTAC were similar (median TBRmax [Interquartile range; IQR]: CTAC = 1.65 [1.23–2.38], RACTAC = 1.63 [1.23–2.33], p =0.55). (Figure 4, A). Good agreement of the TBR measures obtained using the two different attenuation correction techniques was observed with coefficient of reproducibility of 0.18 (Figure 4, B).

Figure 4.

A) Boxplot with connecting lines between the TBRmax measurements using the RACTAC maps vs CTAC maps (blue and red boxes represent interquartile range, with a thick solid line inside represents the median), B) Bland-Altman plot of the differences between the TBRmax measured using RACTAC maps vs CTAC maps.

TBR_max = maximum Target to background ratio, CTAC = Computed Tomography Attenuation Correction and RACTAC = Respiratory averaged computed tomography attenuation correction, CR = coefficient of reproducibility

Per vessel analyses (CMA)

Assessments of the individual vessel microcalcification burden (CMA) revealed no major differences between PET images reconstructed using CTAC and RACTAC scans (median [IQR] CMA: CTAC = 0.10 [0–1.0], RACTAC = 0.15 [0–1.03], p=0.19) (Figure 5, A). Bland-Altman plots of the CMA values revealed a high degree of agreement when comparing the per vessel burden (Figure 5, B), with coefficient of reproducibility of 0.17.

Figure 5.

A) Boxplot with connecting lines between the CMA measurements using the RACTAC maps vs CTAC maps (blue and red boxes represent interquartile range, with thick solid line inside represents the median), B) Bland-Altman plot of the differences between the CMA measured using RACTAC maps vs CTAC maps.

CMA = coronary microcalcification activity, CTAC = Computed Tomography Attenuation Correction and RACTAC = Respiratory averaged computed tomography attenuation correction, CR = coefficient of reproducibility

Coronary tree analyses (CMA)

Whole-coronary tree microcalcification burden (CMA for all three vessels combined), was comparable for the two series of reconstructions (CMAtotal: median [IQR]: CTAC = 1.01 [0–1.83], RACTAC = 1.05 [0–1.95], p=0.42) (Figure 6, A). In concordance with the single-vessel CMA burden, the data reconstructed using the two attenuation correction protocols were in agreement (Figure 6, B), with a coefficient of reproducibility of 0.17.

Figure 6.

A) Boxplot with connecting lines between the CMA total measurements using the RACTAC maps vs CTAC maps (blue and red boxes represent interquartile range, with thick solid line inside represents the median), B) Bland-Altman plot of the differences between the CMAtotal measured using RACTAC maps vs CTAC maps.

CMAtotal = whole coronary tree microcalcification activity, CTAC = Computed Tomography Attenuation Correction and RACTAC = Respiratory averaged computed tomography attenuation correction, CR = coefficient of reproducibility

Correlation of lesion activity according to their location

Of the 34 coronary lesions, 2 were recognized in the left main stem (LMS), 15 in the left anterior descending artery, 8 in the right coronary artery and 9 in the left circumflex artery. There was excellent correlation among all lesions (R²=0.97) and there was no difference between different vessels (Figure 7).

Figure 7.

Correlation plot of SUVmax across the different lesion location. Excellent correlation (R²) shown indendent of the cornary artery involved.

LMS= left main stem, LAD= left anterior descending artery, RCA= right coronary artery and LCx= left circumflex artery, SUVmax= maximum standardized uptake value, CTAC = Computed Tomography Attenuation Correction and RACTAC = Respiratory averaged computed tomography attenuation correction

Discussion

In this study, we evaluated the impact of using standard CTAC maps and RACTAC maps obtained from cine-CT maps through respiratory averaging of CT data obtained in 4D on coronary PET. Our main finding was that using RACTAC did not introduce any significant changes in the quantitative comparisons, both when comparing single lesion activities and vessel and the whole coronary tree microcalcification burden. To our knowledge, this is the first study evaluating RACTAC maps in the assessment of coronary PET/CT.

Quantitative accuracy is of high importance in the assessment of both the singular vulnerable plaque and in the overall assessment of the coronary tree microcalcification burden (2). Previous studies have shown that both inter-reader and inter-scan variabilities are within acceptable ranges(6, 25). However, the quantitative accuracy can be impaired by significant respiratory and patient motion shifts during the acquisition (6, 26). Despite improvements in the test-retest reproducibility and reclassification of singular lesions following the introduction of sophisticated motion correction techniques, the quantitative accuracy might still be impaired by respiratory translations of the PET images during the 30-min long PET acquisitions. Current attenuation correction protocols employ a single-phase CT scan (free breathing, or end-expiratory breathing) (5, 12–15). The use of single-phase CTAC maps may change the observed uptake patterns in the heart, with deviations between 6% in canine models (27), and as much as 35% in human studies (28). In addition to the impact of the respiratory averaged CTAC maps, we also need to account for potential problems between co-registered PET and CTAC maps which have shown to affect myocardial perfusion assessments (8, 9, 11). The impact of misregistration of PET emission data and transmission scans (CTAC maps) have not been evaluated in-depth yet. However, as many CTAC maps are acquired using free breathing protocols, the CTAC maps are likely acquired in an expiratory breathing phase, which is similar to the averaged respiratory position during the PET acquisitions. Several efforts to minimize potential mismatches have been proposed, hereunder the feasibility of scanning at the optimal time during respiration (29–31). Additionally, the sizes of coronary lesions are often of the same magnitude as the PET system resolution, which leaves the attenuation correction issues less dominant than partial volume effects.

Unfortunately, even with perfect breath instructions, it is challenging to state the optimal breath protocol applicable for all patients and even if the CT scan was obtained at the optimal breathing state. Pan et al.(17, 32) demonstrated that potential misalignment caused by different breathing phases during helical CT and PET affects both the quantitative and qualitative accuracy. By utilizing RACTAC for attenuation correction instead of the commonly employed CTAC, the frequency and impact of breathing artefacts were reduced, with improved tumor quantification as a result (16). It was argued that by utilizing, fast-scan cine CT acquisition of 5 seconds it is possible to bring together the temporal resolutions of CT and PET (16, 17, 32).

In this study, we did not observe any significant changes in PET uptake values between the reconstructions obtained using CTAC and RACTAC. This finding contrasts with previous reports (8, 17, 27), in which up to 40% of the false-positive results normalized with the use of respiratory averaged CTAC maps (8). Several reasons could explain this. First, the investigators were measuring the activity of the myocardium and because of the higher diaphragmatic position at end-expiration they observed more misregistration artefacts resulting in decreased emission activity in inferior, inferoseptal, and inferolateral walls and compromised quantitative accuracy and the interpretation of myocardial viability. These artifacts were corrected by using respiratory-gated average attenuation maps. In our study, we concentrated on a focal activity away from the diaphragm and did not observe a difference in coronary ¹⁸F-NaF uptake between the single-phase CTAC and respiratory averaged CTAC. Second, in these studies, the investigators measured the average activity, while we measured maximum activity (TBR_MAX) or activity above a specific threshold (CMA). By measuring maximum activity, we have encountered less variability by potential motion-driven misregistration as our volume of interest had a diameter of 5 mm. Gould et al. (8) showed that it was the transaxial misregistration of >6 mm that frequently caused artefactual defects and therefore by applying a volume of interest of 5 mm and measuring maximal activity we encountered less variability in our measurements. Third, the median offset of the CTAC and RACTAC images were of magnitude of 5.4mm (3D), with only 5 patients having translations of more than 10mm between the CTAC and RACTAC maps. These minor offsets combined with the averaging of the attenuation correction maps result in a relatively small impact of RACTAC images when compared to the CTAC maps. Finally, we used cardiac motion-corrected reconstructions(5) improving our ability to discriminate activity coming from the corresponding plaque lesion and improving our co-registration. We have previously shown that cardiac motion causes attenuation artifact and therefore correcting for cardiac motion improves our ability to detect coronary artery disease on myocardial perfusion scans (33). Such cardiac motion correction was not used in previous RACTAC work with myocardial perfusion scans (34). In particular, the effects of cardiac contraction exceed that of respiration with regards to the displacement of the coronaries (cardiac contraction displaces the coronary arteries 8–26mm during the cardiac cycle, while normal respiration leads to movement of the heart of approximately 6–13mm (35).

The coefficient of reproducibility reported in this study is similar to findings reported in test-retest reproducibility studies where both similar interscan variations (6), and inter-reader variations we observed(25). While these results have focused on the test-retest and inter-reader variability for two different studies, our current study reports variations occurring as a result of the attenuation correction. The main finding of this study is that the variations added to the quantitative assessments of the per-lesion and whole coronary tree microcalcification burden are within the range of the reported variations. Therefore, RACTAC applied to the coronary plaque assessments does not seem to change the quantitative ¹⁸F-NaF PET results when compared to the standard CTAC technique.

One concern in terms of employing the cine-CT in the routine clinical assessment is the added radiation dose, which conflicts with the as low as reasonably achievable (ALARA) principle. The ALARA principle might be compromised in studies of ¹⁸F-NaF studies, as we in this study did not find any diagnostic relevant changes in the lesion assessment using the RACTAC corrected reconstructions. Based on the calculations, the dose-burden given to a standard 70kg man equals 9.1mSv. This dose can be divided into the dose of the PET emission data (6.7mSv) and corresponding low-dose CTAC map (1mSv) and an additional 1.4mSv (15.2% extra dose) obtained for the cine-CT scan. While the difference between RACTAC and the regular CTAC is relatively low, the findings of this study indicate that the increased dose does not improve the quantification of coronary plaque scans using ¹⁸F-NaF and, thus, should be omitted in future studies to comply with ALARA.

Limitations

Our study has limitations. First, our sample is small (23 patients). Despite that, we showed that our reproducibility coefficient is below the previously reported interobserver and interscan variation (25). Second, we report results obtained using cardiac motion correction which have less noise and lower TBR values than the commonly presented end-diastolic imaging reconstruction (1, 19). However, previous studies from our group have shown that both end-diastolic and cardiac motion corrected images can be used for ¹⁸F-NaF coronary uptake assessments (5, 6). Third, a correlation between the volume with active microcalcification and the SUVmax or TBR measurements as the volume of activity is unknown. This limitation, although of clinical importance, is not affected by the attenuation correction, as shown in Figure 7 and, thus, not considered to affect the results of this paper. Fourth, it is a single-center study and we used only scanners from one vendor, thus, vendor-specific variations cannot be ruled out. A bigger study involving multiple centers and readers would be required to confirm our findings. Finally, the results presented in this study apply to the coronary PET imaging only, the effect of using RACTAC maps in other studies relying on absolute SUV measurements such as sarcoid PET/CT will need to be evaluated in future studies.

New Knowledge Gained

Using RACTAC instead of CTAC maps does affect quantitative accuracy of ^18F-NaF PET/CT uptake. This finding is important to ensure the optimal quantitative accuracy in studies utilizing ¹⁸F-NaF PET with standard CTAC maps, such as in Prediction of Recurrent Events With 18F-Fluoride [PREFFIR; NCT02278211] study, where investigators use ¹⁸F-NaF PET/CT as a marker of coronary plaque vulnerability to detect culprit and non-culprit unstable coronary plaques in patients with recent myocardial infarctions to determine the prognostic significance of coronary ¹⁸F-NaF uptake. Based on this finding, given a comparable quantitative accuracy of CTAC and RACTAC maps, both approaches can be used depending on the institutional acquisition protocols.

Conclusion

Applying respiration-averaged CT attenuation correction (RACTAC) maps do not affect the quantification of the coronary lesions as read in fusion PET/CT images. Current protocols utilizing single-shot CTAC maps provide equivalent corrections for the clinical reading of the patients.

Abbreviations

RACTAC

respiration-averaged CT attenuation correction

CTAC

CT attenuation correction

CMA

coronary microcalcification activity

¹⁸F-NaF

¹⁸F-sodium fluoride

PET

positron emission tomography

CTA

computed tomography angiography

ECG-MC

cardiac motion corrected

BC

background blood pool clearance correction

TBR_MAX

maximum target to background ratio

SUV_MAX

maximum standardized uptake value

VOI

volume of interest